MXL skipped the standards process and that may need to change

The Media Exchange Layer (MXL) arrived differently from most broadcast technologies.

Rather than emerging from a standards committee, it was built as an open-source software development kit under the Linux Foundation… code first, ratification later, if at all. Version 1 of the specification was published in March 2026. Vendor adoption is underway. Interoperability testing has begun.

And SMPTE, the organization that has defined professional media standards for more than a century, is paying close attention.

“I think once MXL becomes more stable, and there are less changes to it, there may be an interest to bring it into SMPTE,” said Thomas Bowser Mason, director of standards at SMPTE, in a conversation with NCS at the 2026 NAB Show. “That’s what I hear from a few people — they’re interested in that.”

That interest reflects both what MXL has accomplished and what it has not yet resolved.

The technology addresses a real and specific problem: how software applications running on shared compute infrastructure exchange media with each other, without the overhead of packetizing and rebuilding streams. As the industry moves toward software-defined facilities, that problem is increasingly urgent. But the governance model that produced MXL, one where influence is tied to code contribution rather than committee consensus, raises questions that a standards organization like SMPTE is positioned to take seriously.

Understanding the Media Exchange Layer: A shift in how broadcast software connects

The open-source path and its tradeoffs

MXL’s development under the Linux Foundation was a deliberate choice. The approach compresses the timeline between concept and working implementation, and sidesteps the slower process of building consensus across a standards body before anything is built.

Respondents to a recent Industry Insights roundtable on MXL and the Dynamic Media Facility model noted both the advantages and the implications of that model.

Advertisement

Jan Helgesen, head of products and solutions at Nevion, said the Linux Foundation approach “differs significantly from traditional standards-based approaches,” reducing ambiguity in interpreting specifications and moving faster than conventional methods allow. He also noted that influence over the specification correlates with the volume of code contributed, which “may favor larger organizations with greater resources.”

Ian Wagdin, vice president of technology and innovation at Appear, described the “implement-first” strategy as a significant departure from previous standardization efforts in broadcast, one that accelerated adoption while bypassing the de jure process that gives standards their formal standing.

That standing matters.

SMPTE’s standards are accredited through ANSI, which means they can be referenced in legislation and regulation. An open-source SDK, however widely adopted, does not carry that weight. As MXL moves from pilot deployments toward broader infrastructure adoption, the question of whether it needs that kind of formal grounding becomes more pressing.

Bowser Mason said the current moment is one of observation and readiness. SMPTE is not moving to absorb MXL as it stands, the specification is still evolving and adoption is still early. But the organization is not standing apart from the conversation either.

Industry Insights: The Media Exchange Layer’s role in software-defined production

Where SMPTE already intersects

ST 2110, SMPTE’s IP media transport standard, is already a foundational layer in most discussions about MXL deployment. The two technologies are consistently described by vendors and facility planners as complementary: ST 2110 handles media transport across physical IP networks, while MXL handles media exchange between software applications on shared compute. They operate at different layers and are designed to coexist.

That relationship means any formal SMPTE engagement with MXL would not start from scratch. ST 2110 infrastructure is already in place at the facilities where MXL is being piloted, and SMPTE’s existing standards work forms part of the architecture in which MXL sits.

Bowser Mason identified the Dynamic Media Facility model as the larger question for SMPTE.

DMF, the reference architecture being developed through the Joint Taskforce on Dynamic Media Facilities, a working group convened by Advanced Media Workflow Association (AMWA) and the European Broadcasting Union (EBU), describes how software-based media functions should be orchestrated, controlled and secured across a facility. ST 2110 is expected to play a role within that model. So is a control protocol SMPTE has been developing called Catena.

“Catena is really about setting up devices, operating devices, customizing them,” Bowser Mason said. He described it as complementary to AMWA’s NMOS specifications rather than a competitor, with a plug-in mechanism that allows NMOS discovery and registration to connect into the Catena framework.

Catena’s relevance to DMF is practical. As facilities shift from fixed-function hardware to software-defined workflows, the question of how devices and services are configured, addressed and operated across a multi-vendor environment becomes a standards problem. Bowser Mason said multilingual device configuration is one specific area Catena is designed to address — a detail that illustrates how granular the control layer problem becomes at scale.

The DMF complexity question

Roundtable respondents were broadly aligned that DMF, as a concept, is still being defined.

Advertisement

Olivier Suard, vice president of marketing at Nevion, said DMF “is still a concept, with many details still missing.” Adam Marshall, chief product officer at Grass Valley, said the framework needs to keep moving “from discussion to definition.” Russell Trafford-Jones, industry engagement manager at Techex, said the industry needs “mature orchestration, consistent monitoring and proven interoperability between independently procured vendor products” before DMF becomes infrastructure rather than a reference model.

That unfinished state is not unusual for a framework of this scope.

But it does create a specific challenge for an organization like SMPTE, whose standards process is built around stability and precision. The question of when to engage and at which layer requires knowing which parts of the DMF model are sufficiently settled to standardize and which are still in flux.

Bowser Mason acknowledged that tension directly when discussing SMPTE’s broader organizational structure. The organization maintains separate technology communities covering cloud, digital cinema, metadata, compression and other areas. Some of the most significant current challenges, including AI, content provenance and software-defined infrastructure, cut across all of them.

“We’re trying to have these different categories and kind of not keep them apart because some efforts really go across,” he said. “It’s maybe a question for SMPTE — how can we adjust our standards organization in a way that makes that easier? We probably don’t have an idea of what that would look like. I think it’s a discussion we have to have in the next year.”

Fragmentation as a forcing function

Sally-Ann D’Amato, executive director of SMPTE, framed the urgency in terms of what happens without a coordinating body.

In broadcast and media, as in other technology sectors, the risk of parallel and incompatible implementations is real. If vendors develop independent authentication schemes, tagging systems or interoperability frameworks, the result is a loss of the shared reference points that make multi-vendor environments workable.

“If everybody has their own single source of truth or way to tag things and track assets, then there’s really no way to track,” D’Amato said.

That observation applies as directly to MXL’s ecosystem as it does to content provenance.

Roundtable respondents noted that vendor buy-in to MXL’s open-source development is essential for the framework to function as intended. Chris Scheck, head of marketing content at Lawo, said broad vendor contribution “is crucial.” But contributing to an open-source project and adopting a formal standard are different levels of commitment, with different implications for long-term interoperability.

Advertisement

SMPTE’s potential role is not to replace what the Linux Foundation process has produced, but to provide a layer of formal standing that open-source development alone cannot. Whether MXL needs that and when is a question the organization is positioned to help the industry answer.

“Our job is to bring them all together,” D’Amato said, “and have them discuss together and then come to a conclusion, a solution, or at least some guidelines that they can all agree on.”